Speech modelling using cepstral-time feature matrices and hidden Markov models

@inproceedings{citeulike:693465, abstract = {Conventional HMMs assume that speech spectral vectors are uncorrelated. The use of information on the temporal evolution of spectral features, within each state, can improve recognition accuracy and produce a more robust recognition system. The authors present experimental results on improvements in speech recognition using cepstral-time matrix units. Experimental evaluation using a spoken digit data base and a spoken alphabet data base, indicates that the use of cepstral-time matrix features in noisy conditions can provide an improvement in recognition of as much as 20\% in comparison to a conventional spectral vector comprising of cepstral, delta cepstral and delta-delta cepstral features}, author = {Milner, B. P. and Vaseghi, S. V. }, citeulike-article-id = {693465}, journal = {Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference on}, keywords = {asr, dynamic, extract, feature}, pages = {I/601--I/604 vol.1}, posted-at = {2006-06-12 06:42:30}, priority = {3}, title = {Speech modelling using cepstral-time feature matrices and hidden Markov models}, url = {http://ieeexplore.ieee.org/xpls/abs\_all.jsp?arnumber=389222}, volume = {i}, year = {1994} }

TY - CONF ID - citeulike:693465 L3 - citeulike-article-id:693465 TI - Speech modelling using cepstral-time feature matrices and hidden Markov models JF - Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference on VL - i SP - I/601 EP - I/604 vol.1 N2 - Conventional HMMs assume that speech spectral vectors are uncorrelated. The use of information on the temporal evolution of spectral features, within each state, can improve recognition accuracy and produce a more robust recognition system. The authors present experimental results on improvements in speech recognition using cepstral-time matrix units. Experimental evaluation using a spoken digit data base and a spoken alphabet data base, indicates that the use of cepstral-time matrix features in noisy conditions can provide an improvement in recognition of as much as 20% in comparison to a conventional spectral vector comprising of cepstral, delta cepstral and delta-delta cepstral features KW - asr KW - dynamic KW - extract KW - feature AU - Milner, BP AU - Vaseghi, SV PY - 1994/// UR - http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=389222 ER -